Behavior of Stabilized Peat Soils in Unconfined Compression Tests

American J. of Engineering and Applied Sciences 1 (4): 274-279, 2008 ISSN 1941-7020 2008 Science Publications Behavior of Stabilized Peat Soils in Unconfined Compression Tests Wong Leong Sing, Roslan Hashim and Faisal Haji Ali Department of Civil Engineering, Faculty of Engineering, University of Malaya, 50603 Lembah Pantai, Kuala Lumpur, Malaysia Abstract: Problem statement: Deep stabilized peat columns were known to be economical at forming foundations to support highway embankments constructed on deep peat land. However, failure in the formation of the columns with adequate strength was often attributed to unsuitable type and insufficient dosage of binder added to the soil. Organic matter in peat was known to impede the cementing process in the soil, thus retarding the early strength gain of stabilized peat. Approach: To evaluate the strength characteristics of stabilized peat, laboratory investigation on early strength gain of the stabilized soil was conducted to formulate a suitable and economical mix design that could be effectively used for the soil stabilization. To achieve such purpose, the study examined the effect of binder, sodium chloride as cement accelerator and siliceous sand as filler on the unconfined compressive strength of stabilized peat soils after 7 days of curing. Binders used to stabilize the peat were Ordinary Portland cement, ground granulated blast furnace slag, sodium bentonite, kaolinite, lime and bentonite. All the stabilized peat specimens were tested using unconfined compression apparatus. Results: The test results revealed that the stabilized peat specimen (80% OPC: 10% GGBS: 10% SB) with addition of 4% sodium chloride by weight of binder and 50% well graded siliceous sand by volume of wet peat at 300 kg m -3 binder dosage yielded the highest unconfined compressive strength of 196 kpa. Such finding implied that the higher the dosage of siliceous sand in stabilized peat, the more solid particles were available for the binder to unite and form a load sustainable stabilized peat. Conclusions/Recommendations: It could be summarized that as the rate of hydration process of stabilized peat was accelerated by inclusion of sodium chloride, the solid particles contributed to the hardening of stabilized peat by providing the cementation bonds to form between contact points of the particles. Key words: Organic matter, binder, sodium chloride, siliceous sand, unconfined compressive strength INTRODUCTION Peat originates from plant and denotes the various stages in the humification process where the plant structure can be discerned [1]. The decaying process of plant under acidic condition without microbial process results in the formation of organic matter in peat. This renders peat extremely soft and problematic type of soil. It is generally recognized that organic matter tends to retard the hydration and secondary pozzolanic reactions of peat stabilized with Ordinary Portland cement. This is possible due to the fact that black humic acid (a component of organic matter) tends to react with calcium liberated from cement hydrolysis to form insoluble calcium humic acid making it difficult for calcium crystallization, which is responsible for the increase of cement soil strength to take place [2]. Secondary pozzolanic reaction of cement stabilized peat is retarded due to insufficient silica (SiO 2 ) and alumina (Al 2 O 3 ) that can react with calcium hydroxide [(Ca(OH) 2 ] generated from cement hydration to form secondary calcium silicates and aluminates, which are responsible for the long term strength gain of stabilized peat [3]. To counteract the organic matter, sufficient cement with a small additive of sodium chloride (NaCI) can be added to saturated peat to neutralize the black humic acid and provide adequate tricalcium silicate (Ca 3 SiO 5 ) that reacts with water (H 2 O) to produce enough calcium silicate hydrates that bind the soil particles together. In addition, inclusion of well graded siliceous sand increases the packing efficiency of the stabilized peat. In view of that, it is extremely important to ensure that early strength gain of stabilized peat is achieved to offset the strength retarding effect as a result of the Corresponding Author: Wong Leong Sing, Department of Civil Engineering, Faculty of Engineering, University of Malaya, 50603 Lembah Pantai, Kuala Lumpur, Malaysia Tel: +6016 981 8692 Fax: +603 7967 5318 274

presence of acidic organic matter in peat. Since strength of stabilized peat is largely measured by its unconfined compressive strength due to the soil low permeability and high stiffness, unconfined compression tests often provide relatively fast and cheap mean of determining the soil strength. As such, the study focused on evaluating the early unconfined compressive strength gain of peat soil stabilized by different combinations and compositions of Ordinary Portland cement, sodium chloride, siliceous sand, kaolinite, sodium bentonite, bentonite, lime and ground granulated blast furnace slag. The study also investigated the roles of sodium chloride as cement accelerator and well graded siliceous sand as filler at increasing the unconfined compressive strength of stabilized peat. MATERIALS AND METHODS Peat soil sample: Sampling of peat for the study was carried out in Sri Nadi village, Klang, Selangor, Malaysia. For the sampling purpose, excavation of trial pits was done to a depth of 1 m below the ground surface in order to get both undisturbed and disturbed peat samples. Ground water table was found to be about 0.3 m from the ground surface. The high ground water table suggested that the peat had a very high water retention capacity. Visual inspection on the soil indicated that the soil was dark brown in colour. When the soil was squeezed and extruded between fingers, it was found to be somewhat pasty and the plant structure was hardly identifiable. Based on von Post scale on peat s degree of humification, the soil can be classified as H 4. Basic properties of Klang peat are shown in Table 1. From the Table 1, it can be observed that while the peat was highly organic with average organic content of 96%, its average fiber content of 90% implied that the soil was fibrous in nature. According to Dhowian and Edil [4], if peat has 20% fiber content or more, then it can be classified as fibrous peat. Otherwise, the peat is categorized as amorphous peat. Stabilized soil specimen preparation: A total of 15 stabilized peat specimens of different mix design were prepared and cured for 7 days before testing. For soil homogenization, the naturally saturated peat was allowed to pass through 2 mm sieve size. Table 1: Basic properties of Klang peat [5] Basic soil property Average Natural moisture content (%) 668 Specific gravity 1.40 Fiber content (%) 90 Organic content (%) 96 Ash content (%) 4 ph of peat 3.51 Peat soil classification Fibrous Am. J. Engg. & Applied Sci., 1 (4): 274-279, 2008 275 Materials larger than 2 mm sieve size like isolated roots, large fibers and stones were removed through the sieving process. A kitchen mixer was then used to initially mix the wet peat to ensure that moisture was uniformly distributed throughout the soil. Next, the soil was mixed with binder and siliceous sand for 10 min before it was filled and tamped in four layers up to 180 mm height in a plastic tube of 50 mm internal diameter and 250 mm height. Each plastic tube was arranged vertically in a rack and submerged in a water tank before the specimen in each tube was allowed to be cured under a pressure of 100 kpa. While the required dosage of binder for each specimen was measured in 300 kg m 3 relative to the weight of wet peat at its natural moisture content of 668% in the plastic tube, the needed amount of sodium chloride was quantified as 4% by weight of binder used for stabilizing the soil. Amount of siliceous sand added to a stabilized peat specimen was either 25 or 50% by volume of wet peat depending on the objective of testing. After curing, each stabilized peat specimen was extruded from the plastic tube and trimmed to a height of 100 mm for testing. Method of testing: After 7 days of curing, unconfined compression tests were performed on all the stabilized peat specimens to determine their unconfined compressive strength. The test was also done on an undisturbed peat specimen so that the unconfined compressive strength of the soil can be compared to those of stabilized ones. All the tests were done in accordance to US ASTM standard [6]. The various compositions of binder and siliceous sand are shown in Table 2. Soil specimen 1 in Table 2 is undisturbed peat specimen. Soil specimens 2 to 5 are stabilized peat specimens with different compositions and types of binder without sodium chloride at a binder dosage of 300 kg m 3 and 25% well graded siliceous sand by volume of wet peat. Four percent sodium chloride by weight of binder was added to the next seven compositions of stabilized peat specimens (soil specimens 6-12) at the same dosage of binder and siliceous sand in order to observe the increase in strength of the stabilized soils as a result of sodium chloride functioning as a cement accelerator to increase the rate of cement setting time, thereby accelerating the rate of hydration reaction between cement and pore water to generate more calcium silicate hydrate crystals in the stabilized soils. Finally, the last three specimens of stabilized peat (soil specimens 13-15) consist of similar dosage of binder and sodium chloride but the dosage of siliceous sand added to each specimen was

Table 2: Untreated peat specimen and the various compositions of binder and siliceous sand of stabilized peat specimens at a binder dosage of 300 kg m 3 in unconfined compression tests Soil specimen No. Composition of soil specimen 1 Undisturbed peat 2 100% OPC 3 75% OPC: 25% GGBS 4 85% OPC: 15% SB 5 80% OPC: 10% GGBS: 10% SB 6 75% OPC: 25% GGBS (4% SC by weight of binder) 7 85% OPC: 15% SB (4% SC by weight of binder) 8 80% OPC: 10% GGBS: 10% SB (4% SC by weight of binder) 9 10% L: 70% OPC: 10% GGBS: 10% SB (4% SC by weight of binder) 10 85% OPC: 15% K (4% SC by weight of binder) 11 85% OPC: 15% B (4% SC by weight of binder) 12 80% OPC: 10% GGBS: 10% B (4% SC by weight of binder) 13 75% OPC: 25% GGBS (4% SC by weight of binder and 50% siliceous sand by volume of peat at 668% moisture content) 14 85% OPC: 15% SB (4% SC by weight of binder and 50% siliceous sand by volume of peat at 668% moisture content) 15 80% OPC: 10% GGBS: 10% SB (4% SC by weight of binder binder and 50% siliceous sand by volume of peat at 668% moisture content) OPC = Ordinary Portland Cement; GGBS = Ground Granulated Blast Furnace Slag; SB = Sodium Bentonite; L = Lime; K = Kaolinite; B = Bentonite; SC = Sodium Chloride (Unless specified in the table, the dosage of siliceous sand for a stabilized peat specimen is 25% by volume of peat at 668.3% moisture content) Fig. 1: Unconfined compression test using 4.5 kn load frame increased to 50% by volume of wet peat in order to investigate the role of siliceous sand as filler at increasing the number of solid particles in it, thereby contributing to its higher strength in comparison to that with less siliceous sand. Basically, inclusion of the filler produces no chemical reaction but it enhances the strength of the stabilized peat by increasing the number of soil particles available for the binders to unite and form a load sustainable stabilized soil structure [3]. Principle of unconfined compression test: In unconfined compression test, a cylindrical soil specimen of 50 mm diameter and 100 mm height is placed between the top and lower platens as shown in Fig. 1. An axial load is then applied vertically in such a way the constant rate of strain of the soil specimen is 1.5 mm min 1. In other words, 1.5% vertical strain is 276 applied on the specimen at every minute. This rate is so rapid relative to the drainage of the specimen that there is no time for significant volume change in spite of the absence of a membrane to seal the specimen [7]. The unconfined compressive strength is taken as the peak stress of the soil stress-strain curve or if no peak stress is identified after the vertical strain reaches 20%, the strength is considered as the peak stress at its 20% corresponding vertical strain. Failure of soil specimen in the test normally implies the condition in which the specimen can sustain no further increase in stress, i.e., the point at which it offers its maximum resistance to deformation in term of axial stress [6]. RESULTS Figure 2 shows the relationship between unconfined compressive stress and vertical strain of stabilized peat specimens of different binder compositions in comparison to that of undisturbed peat specimen. At its natural state, the unconfined compressive strength of undisturbed peat was found to be 18 kpa. After stabilization of 7 curing days at a binder dosage of 300 kg m 3 and 25% siliceous sand by volume of wet peat, peat mixed with 75% OPC and 25% GGBS in dosage composition did not show improvement in unconfined compressive strength when the strength was found to be 17 kpa only. At the same curing time and dosage of binder and siliceous sand, stabilized peat with compositions of 80% OPC: 10% GGBS: 10% SB and 85% OPC: 15% SB showed only slight improvement in unconfined compressive strength when both specimens yielded the strength of 19 and 24 kpa respectively.

Fig. 2: Unconfined compressive stress-vertical strain relationship of stabilized peat specimens at binder dosage of 300 kg m 3 and 25% siliceous sand by volume of wet peat without sodium chloride in comparison to that of undisturbed peat specimen Fig. 3: Relationship between unconfined compressive stress and vertical strain of stabilized peat specimens at binder dosage of 300 kg m 3, 25% siliceous sand by volume of wet peat and 4% sodium chloride by weight of binder A significant improvement in the strength could only be observed when the peat was stabilized with 100% OPC at the same curing time and dosage of binder and siliceous sand. The modulus of elasticity and unconfined compressive strength of the stabilized peat reached 2668 and 63 kpa respectively. The relationship between unconfined compressive strength and vertical strain of stablized peat specimens with each specimen added 4% sodium chloride by weight of binder is shown in Fig. 3. It could be observed from the Fig. 3 that at 300 kg m 3 binder dosage and 25% siliceous sand by volume of wet peat, sodium chloride had no significant effect on the early strength gain of peat soil stabilized at binder compositions of 75% OPC: 25% GGBS, 85% OPC: 15% SB, 80% OPC: 10% GGBS: 10% SB, 85% OPC: 15% B and 80% OPC: 10% GGBS: 10% B. The specimens yielded unconfined compressive strength of 12, 27, 28, 27 and 14 kpa respectively, which were relatively low. However, at the same dosage of binder, siliceous sand and sodium chloride, stabilization of peat with binder compositions of 85% OPC: 15% K and 10% L: 70% OPC: 10% GGBS: 10% SB resulted in higher unconfined compressive strength when the strength of the specimens were found to be 61 and 50 kpa respectively. Figure 4 shows the unconfined compressive stressvertical strain relationship of stabilized peat specimens, each at a binder dosage of 300 kg m 3, 50% siliceous sand by volume of wet peat and 4% sodium chloride by weight of binder. In general, all of the stabilized peat specimens showed markedly improvement in unconfined compressive strength when compared to that of undisturbed peat specimen. 277 Fig. 4: Relationship between unconfined compressive stress and vertical strain of stabilized peat specimens at binder dosage of 300 kg m 3, 50% siliceous sand by volume of wet peat and 4% sodium chloride by weight of binder At the dosage of binder, siliceous sand and sodium chloride, a binder composition of 80% OPC: 10% GGBS: 10% SB yielded the highest unconfined compressive strength of 196 kpa followed by binder compositions of 85% OPC: 15% SB and 75% OPC: 25% with the strength of 194 and 100 kpa respectively. Modulus of elasticity in the specimens of all the three binder compositions sharply increased with the soil parameter values reached 9668, 9664 and 8064 kpa respectively when compared to that of undisturbed peat which was 362 kpa only. The abrupt increase in the strength of the specimens implied that their vertical strains at failure were relatively short, which were 2.33, 2.33 and 1.60% respectively, when compared to that of undisturbed peat, which was 8.00%. This corresponded to the brittle behaviour of the specimens as the stiffer a

specimen was, the shorter was its corresponding vertical strain at failure and the higher the specimen unconfined compressive strength. DISCUSSION Influence of binder composition on the unconfined compressive strength of stabilized peat specimens without sodium chloride: While unconfined compressive strength improvement was reasonably achieved when the peat was mixed with 100% OPC in the dosage composition, no significant early strength gain could be observed in the rest of the dosage compositions of stabilized peat in Fig. 2. This could be explained by the fact that when insufficient cement was added to saturated peat, provision of tricalcium silicate to neutralize the black humic acid was inadequate to induce cement hydration in the soil. Thus, the strength gain of the soil was retarded even with the presence of pozzolan like sodium bentonite. The impediment of cement hydration within the soil also implied that inadequate calcium hydroxide were generated to react with silica and alumina in pozzolan to form secondary calcium silicate hydrate crystals, which were needed for long term strength gain of stabilized peat. added with sodium chloride should be carried out in order to further understand response of a typical pozzolan to the cement and sodium chloride in saturated peat. Impact of siliceous sand addition on the modulus of elasticity, vertical strain, and unconfined compressive strength of stabilized peat specimens with sodium chloride: It was evident from Fig. 4 that apart from sodium chloride as a cement accelerator and pozzolans as the contributing ingredients towards stronger stabilized peat, the high strength of the stabilized soil was also attributed to the addition of siliceous sand. It appeared that the more well graded siliceous sand added to the soil admixture, the more solid particles were present and the more cementation bonds were formed at the contact points between the solid particles, thus interlocking the organic coarse and soil particles in peat to form stabilized peat structure. Furthermore, the better graded the grain distribution of the sand, the smaller the voids and the greater the number and the larger the interparticle contact surfaces, the stronger the effect of cementation [8]. CONCLUSION Effect of sodium chloride on the unconfined compressive strength of stabilized peat specimens: With the highest early strength gain, it was proven that kaolinite in comparison to other pozzolans in Fig. 3 responded well as a pozzolanic material for Ordinary Portland cement with the help of sodium chloride to form stabilized peat soil. It was also evident from the result that the presence of lime in the binder composition of 10% L: 70% OPC: 10% GGBS: 10% SB with sodium chloride actually promoted the cement hydration and secondary pozzolanic reaction process. While addition of lime and cement into the wet peat basically produced more calcium hydroxide to react with slag and sodium bentonite in the stabilized soil admixture, inclusion of small amount of sodium chloride accelerated the rate of cement hydrolysis in saturated peat, thereby speeding up the setting time and increasing the early strength gain of the stabilized soil. The findings revealed that beside siliceous sand acting as filler, reactivity of sodium chloride as a cement accelerator in saturated peat was rather pozzolan specific. Therefore, at the binder dosage and 25% siliceous sand by volume of wet peat, not all pozzolans could react with cement having sodium chloride as an accelerator to produce stabilized peat with high early strength. As such, further research on the chemical reactivity of type of pozzolan with cement 278 Based on the study of the behavior of stabilized peat soils in unconfined compression tests, the following conclusions are drawn. At a dosage of 300 kg m 3 binder and 25% siliceous sand by weight of wet peat, specimen with a binder composition of 100% OPC gave the highest unconfined compressive strength of 63 kpa in comparison to those of the rest of binder compositions tested without sodium chloride after 7 curing days. This is because the amount of cement was adequate to provide enough tricalcium silicate in saturated peat to neutralize the black humic acid, thus promoting cement hydration and increasing the early strength gain of the stabilized soil specimen With addition of 4% sodium chloride by weight of binder at the same dosage of binder, silceous sand and curing days, the reactivity of wet peat to a binder composition is rather pozzolan specific. This is evident when a specimen with a binder composition of 85% OPC: 15% K yielded a significant unconfined compressive strength of 61 kpa in the presence of sodium chloride in comparison to those of the rest of specimens with sodium chloride

At the same dosage of binder but with addition of 50% siliceous sand by mass of wet peat, all the specimens yielded high unconfined compressive strength. The high early strength gain of the specimens was largely contributed by the well graded solid particles in siliceous sand that enabled cementation bonds to be formed at contact points between the solid particles, thus binding the organic and soil particles in saturated peat to form a stabilized peat structure ACKNOWLEDGEMENT Major funding of the study came from Postgraduate Research Fund PS016-2007B under University of Malaya Research University Grant. Acknowledgement is not complete without thanking Mohiddin Hamzah and Rozita Yusop, University of Malaya Geotechnical laboratory technicians for their kind assistance in laboratory work. REFERENCES 1. Hartlen, J. and W. Wolski, 1996. Embankments on Organic Soils. 1st Edn. Elsevier Science B.V., Amsterdam, Holland. ISBN: 0-444-88273-1, pp: 6 2. Chen, H. and Q. Wang, 2006. The behaviour of soft soil stabilization using cement. Bull. Eng. Geol. Environ., 65: 445-448. Doi: 10.1007/s10064-005-0030-1. 3. Wong, L.S., R. Hashim and F. Ali et al., 2008. Influence of curing under stress on the engineering properties of stabilized peat soil. Proceedings of the International Conference on Construction and Building Technology, June 16-20, Grand Seasons Hotel, Kuala Lumpur, Malaysia, pp: 159-179. 4. Dhowian, A.W. and T.B. Edil, 1980. Consolidation behavior of peats. Geotech. Test. J., 3: 105-114. DOI: 10.1520/GTJ10881J. 5. Wong, L.S., R. Hashim and F. Ali et al., 2008. Compression rates of untreated and stabilized peat soils. Elect. J. Geotech. Eng., 13: 1-13. URL: www.ejge.com/2008/ppr0862/ppr0862.pdf. 6. Head, K.H., 1994. Manual of Soil Laboratory Testing. Vol. 2: Permeability, Shear Strength and Compressibility Tests. 2nd Edn., Pentech Press Limited, London, UK., ISBN: 0-7273-1319-3, pp: 582. 7. Terzaghi, K., R.B. Peck, G. Mesri, et al., 1996. Soil Mechanics in Engineering Practice. 3rd Edn. John Wiley and Sons, New York, USA. ISBN: 978-0-471-08658-1, pp: 127. 8. Kezdi, A., 1979. Stabilized Earth Roads. 1st Edn., Akademiai Kiado, Budapest, Hungary. ISBN: 0-444-99786-5, pp:110 279